Improved recognition of control chart patterns using artificial neural networks

Recognition of abnormal patterns in control charts provides clues to reveal potential quality problems in the manufacturing processes. One potentially popular approach for recognizing different control chart patterns (CCPs) is to develop heuristics based on various shape features of the patterns. The advantage of this approach is that the users can easily understand how a particular pattern is identified. However, consistency in the recognition performance is found to be considerably poor in the heuristics approach. Since shape features represent the main characteristics of the patterns in a condensed form, artificial neural network (ANN) with features extracted from the process data as input vector representation can facilitate efficient pattern recognition with a smaller network size. In this paper, a set of seven shape features is selected, whose magnitudes are independent of the process mean and standard deviation under a special representation of the sampling interval in the control chart plot. Based on these features, the CCPs are recognized using a multilayered perceptron neural network trained by back-propagation algorithm. The recognizer can recognize all the eight commonly observed CCPs. Extensive performance evaluation of this recognizer is carried out using simulated pattern data. Numerical results indicate that the developed ANN recognizer can perform well in real time process control applications with respect to both recognition accuracy and consistency.     

Free volume,adhesion, and viscoelastic properties of model nanostructured pressure‐sensitive adhesive based on stoichiometric complex of poly(N‐vinyl pyrrolidone) and poly(ethylene glycol) of disparate chain lengths

Positron annihilation lifetime spectroscopy was used to characterize the size and content of subnanoscopic free volume in a model pressure‐sensitive adhesive based on a stoichiometric hydrogen‐bonded network complex of poly(N‐vinyl pyrrolidone) (PVP) and oligomeric poly(ethylene glycol) (PEG). Adhesive properties were examined with peel and probe tack tests, and mechanical properties were studied with tensile test. Nonequimolar stoichiometry and the structure of PVP–PEG model pressure‐sensitive adhesive blends were found to be determined by the length of PEG short chains. The size and number density of free volume domains in the PVP–PEG blends were determined as functions of blend composition and relative humidity of the surrounding atmosphere, which controls the amount of absorbed water. Correlating the free volume, adhesion behaviors, and tensile properties of the blends, the range of free volume favoring pressure‐sensitive adhesion in examined compositions was established. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Optimization of supercritical fluid extraction of lycopene from tomato skin with central composite rotatable design model

Response surface methodology using the central composite rotatetable design (CCRD) model was used to optimize parameters for supercritical carbon dioxide extraction of lycopene (C₄₀H₅₆) from dried tomato skin. The CCRD consisting of three-factored factorial design with two levels was used in this study. The factors used were temperature of the extraction chamber (40 and 70 °C), pressure of the extraction fluid (25 and 45 MPa), and modifier concentration (5 and 15%). Judged by the lack-of-fit-test, coefficient of determination and the standard errors results from the analysis of variance have shown the model to be adequate. The linear, quadratic and cross-effects were 0.58, 0.28 and 0.05, respectively. The independent variables have significantly (p < 0.05) influenced the extraction of all trans-lycopene from tomato skin. Although, no significant (p > 0.05) individual effect of modifier concentration shown, a synergetic effect was observed. A second-degree polynomial equation was developed from a response surface analysis for all trans-lycopene yield and the highest yield was predicted at 62 °C, 45 MPa (450 bar) and 14% temperature, pressure and modifier concentration, respectively and the recovery of all trans-lycopene was 33%.     

Stability analysis of BWR nuclear-coupled thermal-hydraulics using a simple model

A simple mathematical model is developed to describe the dynamics of the nuclear-coupled thermal-hydraulics in a boiling water reactor (BWR) core. The model, which incorporates the essential features of neutron kinetics and single-phase and two-phase thermal-hydraulics, leads to a simple dynamical system comprised of a set of nonlinear ordinary differential equations (ODEs). The stability boundary is determined and plotted in the inlet-subcooling-number (enthalpy)/external-reactivity operating parameter plane. The eigenvalues of the Jacobian matrix of the dynamical system also are calculated at various steady-states (fixed points); the results are consistent with those of the direct stability analysis and indicate that a Hopf bifurcation occurs as the stability boundary in the operating parameter plane is crossed. Numerical simulations of the time-dependent, nonlinear ODEs are carried out for selected points in the operating parameter plane to obtain the actual damped and growing oscillations in the neutron number density, the channel inlet flow velocity, and the other phase variables. These indicate that the Hopf bifurcation is subcritical, hence, density wave oscillations with growing amplitude could result from a finite perturbation of the system even when it is being operated in the parameter region thought to be safe, i.e. where the steady-state is stable. Finally, the power-flow map, frequently used by reactor operators during start-up and shut-down operation of a BWR, is mapped to the inlet-subcooling-number/neutron-density (operating-parameter/phase-variable) plane, and then related to the stability boundaries for different fixed inlet velocities corresponding to selected points on the flow-control line. Also, the stability boundaries for different fixed inlet subcooling numbers corresponding to those selected points, are plotted in the neutron-density/inlet-velocity phase variable plane and then the points on the flow-control line are related to their respective stability boundaries in this plane. The relationship of the operating points on the flow-control line to their respective stability boundaries in these two planes provides insight into the instability observed in BWRs during low-flow/high-power operating conditions. It also shows that the normal operating point of a BWR is very stable in comparison with other possible operating points on the power-flow map.     

The study of a maintenance float model with Burr failure distribution

This paper considers an operational system made up of N independent and identical units. These units are required to be in service and are supported by a maintenance float. Here, an attempt has been made to model and analyse the maintenance float problem for a Burr failure distribution. For the sake of simplicity, the repair times are assumed to follow exponential and lognormal distributions.     

GERT analysis of sampling plan for system reliability

In this paper the Graphical Evaluation and Review Technique (GERT) has been applied to modelling of an acceptance sampling plan for a two component parallel system exhibiting standby redundancy.     

CSCC for mean of an inverse Gaussian distribution under type I censoring

In this paper Cumulative Sum Control Chart (CSCC) have been constructed for the mean of Inverse Gaussian distributed life test data. The parameters of the V-mask have been determined and finally the expression for the Average Run Length has been derived and illustrated numerically.     

Bayesian CAR models for syndromic surveillance on multiple data streams: Theory and practice

Syndromic surveillance has, so far, considered only simple models for Bayesian inference. This paper details the methodology for a serious, scalable solution to the problem of combining symptom data from a network of US hospitals for early detection of disease outbreaks. The approach requires high-end Bayesian modeling and significant computation, but the strategy described in this paper appears to be feasible and offers attractive advantages over the methods that are currently used in this area. The method is illustrated by application to ten quarters worth of data on opioid drug abuse surveillance from 636 reporting centers, and then compared to two other syndromic surveillance methods using simulation to create known signal in the drug abuse database.